[JDEV] look for help about unicode in jabber system

Andreas Ames andreas.ames at Tenovis.com
Mon Aug 19 04:20:53 CDT 2002


ÕÅ Æé <jabberjaist at hotmail.com> writes:

> =816=A61We need to find some kind of encoding algorithm/API which
> convert= s Unicode to UTF-8 before we send out strings to the server
> and some kind of decodi= ng Algorithm/API which does the opposite
> when we receive strings.

I'm told that the jabber.org jabberd supports UTF-8 exclusively, while
the Jabber Inc. server handles UTF-16 additionaly (on the wire).  A
conversion library is libiconv (http://www.gnu.org/software/libiconv/)
which can convert between lots of encodings.

For something that I would consider a good intro about this subject,
look at http://www.debian.org/doc/manuals/intro-i18n/.  The author
states that the easiest choice in new software packages is to use
wchar_t and locales if you want to avoid support for only one or only
a few encodings.  I wonder why so many projects, including jabberd,
still use UTF-8 encoding _internally_.  It can be a pain, for example
when iterating over strings.  The only reason I can imagine is that
UTF-8 encoding is ASCII-'compatible', which is admittedly a nice
feature when dealing with ASCII-only convenience libs.

About fonts (rendering) for your prefered glyphs I can't help, sorry.



More information about the JDev mailing list